1 |
On 03/13/2012 10:05 PM, Zac Medico wrote: |
2 |
> On 03/13/2012 06:42 PM, Brian Harring wrote: |
3 |
>> Leaving it such that the PM has to enforce things like "don't have |
4 |
>> multiple EAPI assignments" means by default, one of them isn't going |
5 |
>> to... leading to the ebuilds breaking... specifically the common case |
6 |
>> being the ebuild becoming acclimated to some quirk of portage. |
7 |
> |
8 |
> My intention is for PMS to specify the search algorithm that's used to |
9 |
> probe the EAPI, and also for it to specify that package managers must |
10 |
> treat an ebuild as invalid if the probed EAPI is not identical to the |
11 |
> one that's obtained from bash. If all package managers adhere strictly |
12 |
> to these two requirements, then we won't have any incompatibilities |
13 |
> between package managers here. |
14 |
|
15 |
Someone should really throw up a table on wiki.g.o with a comparison of |
16 |
the proposed methods. IIRC, the pros/cons of this in contrast to GLEP 55 |
17 |
are something like, |
18 |
|
19 |
Pro: |
20 |
|
21 |
* We don't need to change the filename, and "ebuild" is nice |
22 |
* GLEP 55 pissed people off, and was already rejected |
23 |
* Some people think the EAPI rightfully belongs in the ebuild |
24 |
|
25 |
Cons: |
26 |
|
27 |
* New features can't be implemented immediately because PMs |
28 |
have to catch up first. |
29 |
* Slight performance hit |
30 |
* Old package managers on out-of-date systems will barf on it |
31 |
* It involves using a magic identifier, e.g. a comment. Magic is |
32 |
bad, and the fact that messing with a comment can break your PM |
33 |
is counter-intuitive. |
34 |
* Some people think the EAPI rightfully belongs in the filename |
35 |
|
36 |
|
37 |
and the last one is worth the most points to everyone anyway. |